Adipose Tissue Insulin Resistance Is Not Associated With Changes in the Degree of Obesity in Children and Adolescents

Abstract

Context: The "carbohydrate-insulin model" claims that adipose tissue insulin sensitivity explains development of obesity via adipocyte energy storage and/or low postprandial metabolic fuel levels.

Objective: We tested whether adipose tissue insulin sensitivity predicts changes in the degree of obesity over time.

Methods: This secondary analysis of an observational study of youth with obesity included 213 youths at a pediatric weight management clinic. Adipose tissue insulin sensitivity/resistance and whole-body insulin sensitivity were evaluated using oral glucose tolerance test (OGTT)-derived surrogates in the face of changes in the degree of obesity over time. The main outcome measure was change in body mass index (BMI) z score.

Results: Mean BMI z change was 0.05 ± 0.28 (range, -1.15 to 1.19), representing a broad distribution of changes in the degree of obesity over a follow-up period of 1.88 ± 1.27 years. Adipose tissue insulin resistance was not associated with changes in the degree of obesity in univariate or multivariate analyses (adjusted for baseline age, BMI z score, sex, ethnicity, and time of follow-up). Low postprandial free fatty acid concentrations or their suppression during the OGTT were not associated with changes in the degree of obesity in univariate or multivariate analyses. Whole-body insulin sensitivity was not associated with changes in the degree of obesity in univariate or multivariate analyses.

Conclusion: In this secondary analysis, in youth with obesity, adipose tissue insulin resistance is not protective from increases of the degree of obesity and skeletal muscle insulin resistance is not associated with increases of the degree of obesity.The analysis was performed using data derived from NCT00000112 and NCT00536250.

Keywords: adipose insulin sensitivity; carbohydrate-insulin model; insulin resistance.

Figure 1.

BMI z score change by Adipo IR1, Adipo IS1, and prandial Adipo IR1 tertiles. Panels a-c are results of univariate comparisons. Adjusted values (panels d-f) are derived from the linear regression models. Values presented as means ± standard errors. The values for Adipo IR1 tertiles are: tertile 1 < 12.88; 12.88 ≤ tertile 2 ≤ 22.59; tertile 3 > 22.59 (Adipo IR1 expressed in nmol/L × μU/mL). The values for Adipo IS1 tertiles are: tertile 1 < 0.08; 0.08 ≤ tertile 2 ≤ 0.14; tertile 3 > 0.14 (Adipo IS1 expressed in 1/(nmol/L × μU/Ml)).

Figure 2.

BMI Z score changes by AUC_FFA and FFA suppression tertiles. Panels a, b are results of univariate comparisons. Adjusted values (panels c, d) are derived from the linear regression models. Values presented as means ± standard errors. The values for AUC FFA tertiles are: tertile 1 < 18.97; 18.97 ≤ tertile 2 ≤ 27.5; tertile 3 > 27.5 (AUC FFA expressed in mmol *min/L). The values for FFA suppression tertiles are: tertile 1 < 9.3; 9.3 ≤ tertile 2 ≤ 15; tertile 3 > 15 (FFA suppression expressed in 1/(nmol/L × μU/Ml)).

Figure 3.

BMI Z score changes by WBISI tertiles. Panel a is result of univariate comparisons. Adjusted values (panel b) are derived from the linear regression model. Values presented as means ± standard errors. The values for WBISI tertiles are: tertile 1 < 1.21; 1.21 ≤ tertile 2 ≤ 1.97; tertile 3 > 1.97.